chore: [PAYMCLOUD-174] Update alert descriptions and configurations (#…

…2570)

Update alert descriptions and configurations

Revised the alert descriptions to be more consistent with module names. Adjusted alert severity, window durations, and evaluation frequencies to improve monitoring accuracy and responsiveness.

Signed-off-by: Fabio Felici <[email protected]>

src/aks-platform/99_locals.tf

@@ -34,17 +34,17 @@ locals {
   aks_logs_alerts = {
     pods_failed = {
       display_name            = "${module.aks.name}-POD-FAILED"
-      description             = "Detect if there is any pod failed"
+      description             = "${module.aks.name} POD FAILED"
       query                   = <<-KQL
         KubePodInventory
         | where TimeGenerated > ago(15m)
         | where PodStatus == "Failed"
         | project TimeGenerated, ClusterName, Namespace, Name, PodStatus
         | summarize count() by PodStatus, Namespace
       KQL
-      severity                = 1
-      window_duration         = "PT30M"
-      evaluation_frequency    = "PT15M"
+      severity                = 2
+      window_duration         = "PT15M"
+      evaluation_frequency    = "PT5M"
       operator                = "GreaterThan"
       threshold               = 1
       time_aggregation_method = "Average"
@@ -65,7 +65,7 @@ locals {
     }
     pods_ready = {
       display_name                             = "${module.aks.name}-POD-READY"
-      description                              = "Detect pods percentage is over defined threshold"
+      description                              = "${module.aks.name} POD Ready under threshold"
       query                                    = <<-KQL
         KubePodInventory
         | where TimeGenerated > ago(15m)
@@ -91,7 +91,7 @@ locals {
     }
     pods_oomkilled = {
       display_name            = "${module.aks.name}-POD-OMMKILLED"
-      description             = "Detect if any pod is OOMKilled"
+      description             = "${module.aks.name} POD OOMKilled"
       query                   = <<-KQL
         KubePodInventory
         | where PodStatus != "running"
@@ -131,7 +131,7 @@ locals {
     }
     pods_restart = {
       display_name            = "${module.aks.name}-POD-RESTART-COUNT"
-      description             = "Detect if any pod was restarted abnormally"
+      description             = "${module.aks.name} POD Restarted multiple times"
       query                   = <<-KQL
         KubePodInventory
         | where ContainerRestartCount > 0
@@ -141,8 +141,8 @@ locals {
         | summarize any(RestartCount) by Namespace
       KQL
       severity                = 2
-      window_duration         = "PT30M"
-      evaluation_frequency    = "PT15M"
+      window_duration         = "PT15M"
+      evaluation_frequency    = "PT5M"
       operator                = "GreaterThan"
       threshold               = 5
       time_aggregation_method = "Average"
@@ -165,7 +165,7 @@ locals {
     }
     pods_cpu = {
       display_name            = "${module.aks.name}-POD-CPU-USAGE"
-      description             = "Detect if any pod has High CPU Usage"
+      description             = "${module.aks.name} POD High CPU Usage"
       query                   = <<-KQL
         let endDateTime = now();
         let startDateTime = ago(1h);
@@ -220,7 +220,7 @@ locals {
     }
     pods_memory = {
       display_name            = "${module.aks.name}-POD-MEM-USAGE"
-      description             = "Detect if any pod has High Memory Usage"
+      description             = "${module.aks.name} POD High Memory Usage"
       query                   = <<-KQL
         let endDateTime = now();
         let startDateTime = ago(1h);